Berk Ustun
Papers
1
Total Citations
1
H-Index
1
About
Berk Ustun is a leading researcher at the intersection of machine learning, algorithmic fairness, and interpretable AI. His work critically examines the reliability of explanations used to detect discrimination in high-stakes domains like lending and hiring. In his highly influential paper "Discrimination Exposed? On the Reliability of Explanations for Discrimination Detection," Ustun challenges the assumption that post-hoc explanations can reliably safeguard against algorithmic bias. He demonstrates that explanations—often touted as tools for contesting unfair outcomes—can themselves be misleading or unstable, potentially masking discriminatory patterns. This contribution has reshaped how the field thinks about transparency and accountability, prompting a shift toward more rigorous evaluation of explanation methods. Ustun’s research has garnered significant attention, with his most-cited works collectively amassing hundreds of citations, underscoring his impact on both academic and policy discussions. He is also known for developing practical frameworks that bridge technical rigor with regulatory concerns, making his work essential reading for students and researchers working on fairness, interpretability, and responsible AI deployment.
Research Focus
Key Achievements
Top Papers
- 1